Bryce Winkelman is chief revenue officer at Typeform, which helps companies collect more data with forms, surveys and quizzes designed to be appealing, and put that data to good use. Winkelman leads the company’s global go-to-market organization, including marketing, sales, partnerships, revenue operations and customer success. In this Q&A, he shares what Typeform’s latest research reveals about the gap between AI adoption and actual efficiency in lead generation, and how sales and marketing teams can rethink their workflows to turn customer intent into revenue.
Typeform just released a new data report to understand how marketers are using AI within lead generation processes. What finding surprised you the most?
What stood out most to me was just how wide the gap is between AI adoption and actual efficiency, particularly within a process like lead generation, where AI should theoretically be taking manual work off teams’ plates and helping them act on leads faster. In our survey of marketers, we found that 97% are using AI for at least one inbound lead follow-up activity, but 70% still spend six or more hours every week manually managing inbound leads. We talk so much about AI saving time, but clearly using the technology and actually getting that time back are two very different things.
You see that disconnect in response times, too. Seven in 10 organizations leave at least one-quarter of their inbound leads waiting more than 24 hours for a response. That’s a long time to leave someone in the lurch who has raised their hand and told you they’re interested. To me, the data is a good reminder that adopting AI isn’t the finish line. The question leaders need to ask is, did anything actually get better?
Why are sales and marketing teams still struggling to respond to inbound leads quickly, even with AI at their disposal?
We’re seeing an AI paradox, where we’re using new tools but still have the same unsolved problem. Marketing teams have added AI to their existing processes without necessarily redesigning those processes around it. That means leads are still moving through disconnected systems, waiting on responses and falling through the cracks of manual handoffs. The only difference now is there’s more automation layered on top (and more wasted money).
We can’t even blame this issue on data. 78% say they already collect enough information to personalize a response to every inbound lead, but 65% still struggle to turn that information into action. That means the breakdown is happening after the lead comes in. Before following up, most marketers hand off leads at least twice, and 68% route them through four or more platforms between capture and first response. Every handoff adds time, and every system a lead has to move through creates another opportunity for that momentum to disappear.
That’s especially damaging because when someone engages with your brand, their intent is high and you have a window to act on it. But instead, we’re leaving potential customers “on read,” and marketers say it’s directly contributing to lost revenue (50%), lower conversion rates (49%), more manual work (48%) and longer sales cycles (46%). Disconnected workflows are creating problems for both the customer and the business.
AI can absolutely help, but we need to use it differently. Right now, marketers are using it throughout lead generation, with the top three use cases being to analyze and visualize data, book demos or follow up with leads, and route and prioritize leads. Those all make individual tasks faster, but the biggest problem often lives between those tasks, where one system or person hands the lead to another. Only 28% of marketing teams use AI to trigger workflows for lead routing and follow-up, and just 24% use it to sync information into their CRM.
Until teams rethink the workflow itself, rather than simply adding AI to individual tasks, they’ll continue to see a loss in revenue and customer satisfaction.
What separates teams that are getting real value from AI from those that are simply using more of it?
The teams seeing ROI from AI tools are being deliberate about where they use the technology. Our research found that the highest-performing teams are looking at the entire process and figuring out where AI can actually improve outcomes. For example, they’re using it to automatically route leads and trigger follow-up workflows. The issue is that only about a quarter of marketers are doing that today.
Leaders need to get out of the habit of asking, “Are we using AI?” and start asking, “What can we do now that we couldn’t do before?” Did response times improve? Are more qualified leads reaching sales? Are customers having an easier experience? Those are much more useful measures than how many AI tools you’ve bought or how frequently your team uses them.
The other piece is continuing to check whether the technology is actually delivering what you bought it to do. If your goal was improving speed-to-lead, look at how long it took you to follow up before you introduced AI and how long it takes now. If that number hasn’t changed, that’s a sign to go back to the workflow and figure out where the lead is still getting stuck.
AI should solve a problem you can name and improve an outcome you can measure. Otherwise, you’re measuring adoption, not value.
Does making lead management more efficient for internal teams necessarily create a better experience for the buyer? How can leaders make sure they’re improving both?
Not necessarily, and I think that’s an important distinction. Marketers say they have enough data to personalize every lead, but that’s not turning into real value. Having the data and automating the process doesn’t automatically mean you’ve understood what the buyer wants from that interaction.
You could build an incredibly efficient workflow that responds to every lead in 30 seconds, but if the response is irrelevant or doesn’t answer why that person reached out, you’ve made your process faster without making their experience better. This is where anchoring on outcome metrics like conversion, pipeline and ARR improvements are critically important. If the so-called improvement doesn’t impact the bottom line, it isn’t doing its job.
That’s why continuous research needs to sit alongside workflow data. Your operational metrics can tell you that a response went out quickly, but talking to customers can tell you whether it was actually useful.
This is another area where sales and marketing teams are overlooking the value AI can bring. It can help teams get deeper customer feedback more regularly, ask follow-up questions and find patterns across what people are telling them. That way, you aren’t making decisions based on what you think a buyer wants, you’re making them based on what they’re actually saying. Then it’s up to those teams to use what they learn to improve the systems they’re automating.
If you had one piece of advice for sales and marketing teams still struggling to integrate automation into their workflows, what would it be?
Start by diagnosing the problem, then bring AI in to solve it. I’m seeing too many teams doing the reverse, and that’s how they get stuck with too many tools that don’t actually deliver on what they promised.
I recommend picking one place where your team is losing time or where the customer is experiencing obvious friction, figure out exactly what’s going wrong, and then decide whether automation can fix it. Don’t underestimate the power of getting your hands dirty. The best way to understand how another human is progressing through a journey is to go through it yourself so you can spot the flaws.
The lead journey is a perfect example. If your team spends six+ hours a week manually managing leads, map out what they’re actually doing during those six hours. If a lead moves through four or more platforms before anyone responds, figure out which of those handoffs are necessary and which can be consolidated. Then, automate the part that removes a real bottleneck and measure whether the outcome changed.
And I’d keep the measurement simple: Did we save meaningful time? Did we respond faster? Is conversion improving? Is the pipeline growing? If you can’t point to something that improved, adding AI to the workflow wasn’t really a win.
